Rabbit’s r1 AI assistant, a gadget that once generated considerable buzz during its debut at CES, is back in the spotlight with a series of updates. However, despite the company’s efforts to enhance its functionality, these changes are unlikely to sway critics who have been skeptical of its capabilities from the start.
One of the most notable additions is the introduction of the “beta rabbit” mode, designed to bolster the device’s conversational AI abilities. This update allows the assistant to handle more complex or multi-step instructions, theoretically making it more useful for everyday tasks. Additionally, the AI is now programmed to ask follow-up questions when faced with unclear instructions, a feature aimed at improving user interaction.
For example, users might ask, “beta rabbit, can you suggest three books similar to ‘The Power of Now,’ include page length, year of release, and ratings, and save that as a note titled ‘reading list.’ Also, include pictures of the authors.” Following this, they could further instruct, “beta rabbit, can you also get me summaries for those three books?”
While these enhancements sound impressive on paper, the reality of using Rabbit’s AI assistant for such tasks might not live up to the hype. The device’s ability to set travel itineraries, find deals, or recommend products may showcase its web-scraping skills, but its utility remains questionable, particularly on a device with such a small screen. Moreover, users are unlikely to place much trust in book recommendations that are haphazardly sourced.
The update also includes improvements to basic features like setting alarms and timers. However, even these enhancements aren’t immune to scrutiny. For instance, asking the device to “Set a timer for baking chocolate chip cookies” might lead to more questions than answers. What temperature should the oven be set to? How many cookies are being baked? Without this crucial information, the AI could lead users to culinary mishaps. On the other hand, a simpler request, such as “How long should two dozen chocolate chip cookies go in the oven at 300 degrees?” might yield a more reliable response.
Yet, the most anticipated feature, the so-called “large action model” (LAM) — promised to autonomously navigate phone and web app interfaces to accomplish user-defined tasks — remains frustratingly elusive. This capability, heavily touted during the product’s initial unveiling, has yet to be demonstrated in any meaningful way. So far, any actions attributed to LAM are indistinguishable from what could be achieved with standard APIs or basic action scripting.
Despite these shortcomings, there remains a glimmer of hope for Rabbit’s r1. The potential utility of this quirky gadget keeps some users, including myself, from relegating it to the confines of a drawer. But with each update, the question lingers: will Rabbit’s AI assistant ever deliver on its initial promises, or is it destined to remain a novelty with limited real-world application?
I’ve reached out to Rabbit for further details on the status of the LAM and will update this story should any new information come to light.
